Akkariin/youtube-dl
1
0
Fork 0
mirror of https://github.com/l1ving/youtube-dl synced 2025-03-10 18:57:20 +08:00
Code Issues Releases Wiki Activity

Merge branch 'master' of https://github.com/ytdl-org/youtube-dl

Browse Source 
# Conflicts:
#	youtube_dl/extractor/twitter.py
This commit is contained in:
Avi Peretz 2019-11-14 11:18:15 +02:00
commit 7ed03ea77e
16 changed files with 474 additions and 963 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

95
youtube_dl/extractor/addanime.py
View File

@ -1,95 +0,0 @@
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..compat import (
compat_HTTPError,
compat_str,
compat_urllib_parse_urlencode,
compat_urllib_parse_urlparse,
)
from ..utils import (
ExtractorError,
qualities,
)
class AddAnimeIE(InfoExtractor):
_VALID_URL = r'https?://(?:\w+\.)?add-anime\.net/(?:watch_video\.php\?(?:.*?)v=|video/)(?P<id>[\w_]+)'
_TESTS = [{
'url': 'http://www.add-anime.net/watch_video.php?v=24MR3YO5SAS9',
'md5': '72954ea10bc979ab5e2eb288b21425a0',
'info_dict': {
'id': '24MR3YO5SAS9',
'ext': 'mp4',
'description': 'One Piece 606',
'title': 'One Piece 606',
},
'skip': 'Video is gone',
}, {
'url': 'http://add-anime.net/video/MDUGWYKNGBD8/One-Piece-687',
'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
try:
webpage = self._download_webpage(url, video_id)
except ExtractorError as ee:
if not isinstance(ee.cause, compat_HTTPError) or \
ee.cause.code != 503:
raise
redir_webpage = ee.cause.read().decode('utf-8')
action = self._search_regex(
r'<form id="challenge-form" action="([^"]+)"',
redir_webpage, 'Redirect form')
vc = self._search_regex(
r'<input type="hidden" name="jschl_vc" value="([^"]+)"/>',
redir_webpage, 'redirect vc value')
av = re.search(
r'a\.value = ([0-9]+)[+]([0-9]+)[*]([0-9]+);',
redir_webpage)
if av is None:
raise ExtractorError('Cannot find redirect math task')
av_res = int(av.group(1)) + int(av.group(2)) * int(av.group(3))
parsed_url = compat_urllib_parse_urlparse(url)
av_val = av_res + len(parsed_url.netloc)
confirm_url = (
parsed_url.scheme + '://' + parsed_url.netloc
+ action + '?'
+ compat_urllib_parse_urlencode({
'jschl_vc': vc, 'jschl_answer': compat_str(av_val)}))
self._download_webpage(
confirm_url, video_id,
note='Confirming after redirect')
webpage = self._download_webpage(url, video_id)
FORMATS = ('normal', 'hq')
quality = qualities(FORMATS)
formats = []
for format_id in FORMATS:
rex = r"var %s_video_file = '(.*?)';" % re.escape(format_id)
video_url = self._search_regex(rex, webpage, 'video file URLx',
fatal=False)
if not video_url:
continue
formats.append({
'format_id': format_id,
'url': video_url,
'quality': quality(format_id),
})
self._sort_formats(formats)
video_title = self._og_search_title(webpage)
video_description = self._og_search_description(webpage)
return {
'_type': 'video',
'id': video_id,
'formats': formats,
'title': video_title,
'description': video_description
}

74
youtube_dl/extractor/comcarcoff.py
View File

@ -1,74 +0,0 @@
# coding: utf-8
from __future__ import unicode_literals
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
int_or_none,
parse_duration,
parse_iso8601,
)
class ComCarCoffIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?comediansincarsgettingcoffee\.com/(?P<id>[a-z0-9\-]*)'
_TESTS = [{
'url': 'http://comediansincarsgettingcoffee.com/miranda-sings-happy-thanksgiving-miranda/',
'info_dict': {
'id': '2494164',
'ext': 'mp4',
'upload_date': '20141127',
'timestamp': 1417107600,
'duration': 1232,
'title': 'Happy Thanksgiving Miranda',
'description': 'Jerry Seinfeld and his special guest Miranda Sings cruise around town in search of coffee, complaining and apologizing along the way.',
},
'params': {
'skip_download': 'requires ffmpeg',
}
}]
def _real_extract(self, url):
display_id = self._match_id(url)
if not display_id:
display_id = 'comediansincarsgettingcoffee.com'
webpage = self._download_webpage(url, display_id)
full_data = self._parse_json(
self._search_regex(
r'window\.app\s*=\s*({.+?});\n', webpage, 'full data json'),
display_id)['videoData']
display_id = full_data['activeVideo']['video']
video_data = full_data.get('videos', {}).get(display_id) or full_data['singleshots'][display_id]
video_id = compat_str(video_data['mediaId'])
title = video_data['title']
formats = self._extract_m3u8_formats(
video_data['mediaUrl'], video_id, 'mp4')
self._sort_formats(formats)
thumbnails = [{
'url': video_data['images']['thumb'],
}, {
'url': video_data['images']['poster'],
}]
timestamp = int_or_none(video_data.get('pubDateTime')) or parse_iso8601(
video_data.get('pubDate'))
duration = int_or_none(video_data.get('durationSeconds')) or parse_duration(
video_data.get('duration'))
return {
'id': video_id,
'display_id': display_id,
'title': title,
'description': video_data.get('description'),
'timestamp': timestamp,
'duration': duration,
'thumbnails': thumbnails,
'formats': formats,
'season_number': int_or_none(video_data.get('season')),
'episode_number': int_or_none(video_data.get('episode')),
'webpage_url': 'http://comediansincarsgettingcoffee.com/%s' % (video_data.get('urlSlug', video_data.get('slug'))),
}

2
youtube_dl/extractor/common.py
View File

@ -2689,7 +2689,7 @@ class InfoExtractor(object):
entry = {
'id': this_video_id,
'title': unescapeHTML(video_data['title'] if require_title else video_data.get('title')),
'description': video_data.get('description'),
'description': clean_html(video_data.get('description')),
'thumbnail': urljoin(base_url, self._proto_relative_url(video_data.get('image'))),
'timestamp': int_or_none(video_data.get('pubdate')),
'duration': float_or_none(jwplayer_data.get('duration') or video_data.get('duration')),

154
youtube_dl/extractor/daisuki.py
View File

@ -1,154 +0,0 @@
from __future__ import unicode_literals
import base64
import json
import random
import re
from .common import InfoExtractor
from ..aes import (
aes_cbc_decrypt,
aes_cbc_encrypt,
)
from ..compat import compat_b64decode
from ..utils import (
bytes_to_intlist,
bytes_to_long,
extract_attributes,
ExtractorError,
intlist_to_bytes,
js_to_json,
int_or_none,
long_to_bytes,
pkcs1pad,
)
class DaisukiMottoIE(InfoExtractor):
_VALID_URL = r'https?://motto\.daisuki\.net/framewatch/embed/[^/]+/(?P<id>[0-9a-zA-Z]{3})'
_TEST = {
'url': 'http://motto.daisuki.net/framewatch/embed/embedDRAGONBALLSUPERUniverseSurvivalsaga/V2e/760/428',
'info_dict': {
'id': 'V2e',
'ext': 'mp4',
'title': '#117 SHOWDOWN OF LOVE! ANDROIDS VS UNIVERSE 2!!',
'subtitles': {
'mul': [{
'ext': 'ttml',
}],
},
},
'params': {
'skip_download': True, # AES-encrypted HLS stream
},
}
# The public key in PEM format can be found in clientlibs_anime_watch.min.js
_RSA_KEY = (0xc5524c25e8e14b366b3754940beeb6f96cb7e2feef0b932c7659a0c5c3bf173d602464c2df73d693b513ae06ff1be8f367529ab30bf969c5640522181f2a0c51ea546ae120d3d8d908595e4eff765b389cde080a1ef7f1bbfb07411cc568db73b7f521cedf270cbfbe0ddbc29b1ac9d0f2d8f4359098caffee6d07915020077d, 65537)
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
flashvars = self._parse_json(self._search_regex(
r'(?s)var\s+flashvars\s*=\s*({.+?});', webpage, 'flashvars'),
video_id, transform_source=js_to_json)
iv = [0] * 16
data = {}
for key in ('device_cd', 'mv_id', 'ss1_prm', 'ss2_prm', 'ss3_prm', 'ss_id'):
data[key] = flashvars.get(key, '')
encrypted_rtn = None
# Some AES keys are rejected. Try it with different AES keys
for idx in range(5):
aes_key = [random.randint(0, 254) for _ in range(32)]
padded_aeskey = intlist_to_bytes(pkcs1pad(aes_key, 128))
n, e = self._RSA_KEY
encrypted_aeskey = long_to_bytes(pow(bytes_to_long(padded_aeskey), e, n))
init_data = self._download_json(
'http://motto.daisuki.net/fastAPI/bgn/init/',
video_id, query={
's': flashvars.get('s', ''),
'c': flashvars.get('ss3_prm', ''),
'e': url,
'd': base64.b64encode(intlist_to_bytes(aes_cbc_encrypt(
bytes_to_intlist(json.dumps(data)),
aes_key, iv))).decode('ascii'),
'a': base64.b64encode(encrypted_aeskey).decode('ascii'),
}, note='Downloading JSON metadata' + (' (try #%d)' % (idx + 1) if idx > 0 else ''))
if 'rtn' in init_data:
encrypted_rtn = init_data['rtn']
break
self._sleep(5, video_id)
if encrypted_rtn is None:
raise ExtractorError('Failed to fetch init data')
rtn = self._parse_json(
intlist_to_bytes(aes_cbc_decrypt(bytes_to_intlist(
compat_b64decode(encrypted_rtn)),
aes_key, iv)).decode('utf-8').rstrip('\0'),
video_id)
title = rtn['title_str']
formats = self._extract_m3u8_formats(
rtn['play_url'], video_id, ext='mp4', entry_protocol='m3u8_native')
subtitles = {}
caption_url = rtn.get('caption_url')
if caption_url:
# mul: multiple languages
subtitles['mul'] = [{
'url': caption_url,
'ext': 'ttml',
}]
return {
'id': video_id,
'title': title,
'formats': formats,
'subtitles': subtitles,
}
class DaisukiMottoPlaylistIE(InfoExtractor):
_VALID_URL = r'https?://motto\.daisuki\.net/(?P<id>information)/'
_TEST = {
'url': 'http://motto.daisuki.net/information/',
'info_dict': {
'title': 'DRAGON BALL SUPER',
},
'playlist_mincount': 117,
}
def _real_extract(self, url):
playlist_id = self._match_id(url)
webpage = self._download_webpage(url, playlist_id)
entries = []
for li in re.findall(r'(<li[^>]+?data-product_id="[a-zA-Z0-9]{3}"[^>]+>)', webpage):
attr = extract_attributes(li)
ad_id = attr.get('data-ad_id')
product_id = attr.get('data-product_id')
if ad_id and product_id:
episode_id = attr.get('data-chapter')
entries.append({
'_type': 'url_transparent',
'url': 'http://motto.daisuki.net/framewatch/embed/%s/%s/760/428' % (ad_id, product_id),
'episode_id': episode_id,
'episode_number': int_or_none(episode_id),
'ie_key': 'DaisukiMotto',
})
return self.playlist_result(entries, playlist_title='DRAGON BALL SUPER')

57
youtube_dl/extractor/drtv.py
View File

@ -17,6 +17,7 @@ from ..utils import (
float_or_none,
mimetype2ext,
str_or_none,
try_get,
unified_timestamp,
update_url_query,
url_or_none,
@ -24,7 +25,14 @@ from ..utils import (
class DRTVIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?dr\.dk/(?:tv/se|nyheder|radio(?:/ondemand)?)/(?:[^/]+/)*(?P<id>[\da-z-]+)(?:[/#?]|$)'
_VALID_URL = r'''(?x)
https?://
(?:
(?:www\.)?dr\.dk/(?:tv/se|nyheder|radio(?:/ondemand)?)/(?:[^/]+/)*|
(?:www\.)?(?:dr\.dk|dr-massive\.com)/drtv/(?:se|episode)/
)
(?P<id>[\da-z_-]+)
'''
_GEO_BYPASS = False
_GEO_COUNTRIES = ['DK']
IE_NAME = 'drtv'
@ -83,6 +91,26 @@ class DRTVIE(InfoExtractor):
}, {
'url': 'https://www.dr.dk/radio/p4kbh/regionale-nyheder-kh4/p4-nyheder-2019-06-26-17-30-9',
'only_matching': True,
}, {
'url': 'https://www.dr.dk/drtv/se/bonderoeven_71769',
'info_dict': {
'id': '00951930010',
'ext': 'mp4',
'title': 'Bonderøven (1:8)',
'description': 'md5:3cf18fc0d3b205745d4505f896af8121',
'timestamp': 1546542000,
'upload_date': '20190103',
'duration': 2576.6,
},
'params': {
'skip_download': True,
},
}, {
'url': 'https://www.dr.dk/drtv/episode/bonderoeven_71769',
'only_matching': True,
}, {
'url': 'https://dr-massive.com/drtv/se/bonderoeven_71769',
'only_matching': True,
}]
def _real_extract(self, url):
@ -100,13 +128,32 @@ class DRTVIE(InfoExtractor):
webpage, 'video id', default=None)
if not video_id:
video_id = compat_urllib_parse_unquote(self._search_regex(
video_id = self._search_regex(
r'(urn(?:%3A|:)dr(?:%3A|:)mu(?:%3A|:)programcard(?:%3A|:)[\da-f]+)',
webpage, 'urn'))
webpage, 'urn', default=None)
if video_id:
video_id = compat_urllib_parse_unquote(video_id)
_PROGRAMCARD_BASE = 'https://www.dr.dk/mu-online/api/1.4/programcard'
query = {'expanded': 'true'}
if video_id:
programcard_url = '%s/%s' % (_PROGRAMCARD_BASE, video_id)
else:
programcard_url = _PROGRAMCARD_BASE
page = self._parse_json(
self._search_regex(
r'data\s*=\s*({.+?})\s*(?:;|</script)', webpage,
'data'), '1')['cache']['page']
page = page[list(page.keys())[0]]
item = try_get(
page, (lambda x: x['item'], lambda x: x['entries'][0]['item']),
dict)
video_id = item['customId'].split(':')[-1]
query['productionnumber'] = video_id
data = self._download_json(
'https://www.dr.dk/mu-online/api/1.4/programcard/%s' % video_id,
video_id, 'Downloading video JSON', query={'expanded': 'true'})
programcard_url, video_id, 'Downloading video JSON', query=query)
title = str_or_none(data.get('Title')) or re.sub(
r'\s*\|\s*(?:TV\s*\|\s*DR|DRTV)$', '',

12
youtube_dl/extractor/extractors.py
View File

@ -18,7 +18,6 @@ from .acast import (
ACastIE,
ACastChannelIE,
)
from .addanime import AddAnimeIE
from .adn import ADNIE
from .adobeconnect import AdobeConnectIE
from .adobetv import (
@ -223,7 +222,6 @@ from .comedycentral import (
ComedyCentralTVIE,
ToshIE,
)
from .comcarcoff import ComCarCoffIE
from .commonmistakes import CommonMistakesIE, UnicodeBOMIE
from .commonprotocols import (
MmsIE,
@ -254,10 +252,6 @@ from .dailymotion import (
DailymotionPlaylistIE,
DailymotionUserIE,
)
from .daisuki import (
DaisukiMottoIE,
DaisukiMottoPlaylistIE,
)
from .daum import (
DaumIE,
DaumClipIE,
@ -629,7 +623,6 @@ from .microsoftvirtualacademy import (
MicrosoftVirtualAcademyIE,
MicrosoftVirtualAcademyCourseIE,
)
from .minhateca import MinhatecaIE
from .ministrygrid import MinistryGridIE
from .minoto import MinotoIE
from .miomio import MioMioIE
@ -932,10 +925,6 @@ from .rentv import (
from .restudy import RestudyIE
from .reuters import ReutersIE
from .reverbnation import ReverbNationIE
from .revision3 import (
Revision3EmbedIE,
Revision3IE,
)
from .rice import RICEIE
from .rmcdecouverte import RMCDecouverteIE
from .ro220 import Ro220IE
@ -1241,6 +1230,7 @@ from .twitter import (
TwitterCardIE,
TwitterIE,
TwitterAmplifyIE,
TwitterBroadcastIE,
)
from .udemy import (
UdemyIE,

2
youtube_dl/extractor/facebook.py
View File

@ -356,7 +356,7 @@ class FacebookIE(InfoExtractor):
if not video_data:
server_js_data = self._parse_json(
self._search_regex(
r'bigPipe\.onPageletArrive\(({.+?})\)\s*;\s*}\s*\)\s*,\s*["\']onPageletArrive\s+(?:stream_pagelet|pagelet_group_mall|permalink_video_pagelet)',
r'bigPipe\.onPageletArrive\(({.+?})\)\s*;\s*}\s*\)\s*,\s*["\']onPageletArrive\s+(?:pagelet_group_mall|permalink_video_pagelet|hyperfeed_story_id_\d+)',
webpage, 'js data', default='{}'),
video_id, transform_source=js_to_json, fatal=False)
video_data = extract_from_jsmods_instances(server_js_data)

70
youtube_dl/extractor/minhateca.py
View File

@ -1,70 +0,0 @@
# coding: utf-8
from __future__ import unicode_literals
from .common import InfoExtractor
from ..utils import (
int_or_none,
parse_duration,
parse_filesize,
sanitized_Request,
urlencode_postdata,
)
class MinhatecaIE(InfoExtractor):
_VALID_URL = r'https?://minhateca\.com\.br/[^?#]+,(?P<id>[0-9]+)\.'
_TEST = {
'url': 'http://minhateca.com.br/pereba/misc/youtube-dl+test+video,125848331.mp4(video)',
'info_dict': {
'id': '125848331',
'ext': 'mp4',
'title': 'youtube-dl test video',
'thumbnail': r're:^https?://.*\.jpg$',
'filesize_approx': 1530000,
'duration': 9,
'view_count': int,
}
}
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
token = self._html_search_regex(
r'<input name="__RequestVerificationToken".*?value="([^"]+)"',
webpage, 'request token')
token_data = [
('fileId', video_id),
('__RequestVerificationToken', token),
]
req = sanitized_Request(
'http://minhateca.com.br/action/License/Download',
data=urlencode_postdata(token_data))
req.add_header('Content-Type', 'application/x-www-form-urlencoded')
data = self._download_json(
req, video_id, note='Downloading metadata')
video_url = data['redirectUrl']
title_str = self._html_search_regex(
r'<h1.*?>(.*?)</h1>', webpage, 'title')
title, _, ext = title_str.rpartition('.')
filesize_approx = parse_filesize(self._html_search_regex(
r'<p class="fileSize">(.*?)</p>',
webpage, 'file size approximation', fatal=False))
duration = parse_duration(self._html_search_regex(
r'(?s)<p class="fileLeng[ht][th]">.*?class="bold">(.*?)<',
webpage, 'duration', fatal=False))
view_count = int_or_none(self._html_search_regex(
r'<p class="downloadsCounter">([0-9]+)</p>',
webpage, 'view count', fatal=False))
return {
'id': video_id,
'url': video_url,
'title': title,
'ext': ext,
'filesize_approx': filesize_approx,
'duration': duration,
'view_count': view_count,
'thumbnail': self._og_search_thumbnail(webpage),
}

14
youtube_dl/extractor/nexx.py
View File

@ -108,7 +108,7 @@ class NexxIE(InfoExtractor):
@staticmethod
def _extract_domain_id(webpage):
mobj = re.search(
r'<script\b[^>]+\bsrc=["\'](?:https?:)?//require\.nexx(?:\.cloud|cdn\.com)/(?P<id>\d+)',
r'<script\b[^>]+\bsrc=["\'](?:https?:)?//(?:require|arc)\.nexx(?:\.cloud|cdn\.com)/(?:sdk/)?(?P<id>\d+)',
webpage)
return mobj.group('id') if mobj else None
@ -123,7 +123,7 @@ class NexxIE(InfoExtractor):
domain_id = NexxIE._extract_domain_id(webpage)
if domain_id:
for video_id in re.findall(
r'(?is)onPLAYReady.+?_play\.init\s*\(.+?\s*,\s*["\']?(\d+)',
r'(?is)onPLAYReady.+?_play\.(?:init|(?:control\.)?addPlayer)\s*\(.+?\s*,\s*["\']?(\d+)',
webpage):
entries.append(
'https://api.nexx.cloud/v3/%s/videos/byid/%s'
@ -410,8 +410,8 @@ class NexxIE(InfoExtractor):
class NexxEmbedIE(InfoExtractor):
_VALID_URL = r'https?://embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?P<id>[^/?#&]+)'
_TEST = {
_VALID_URL = r'https?://embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?:video/)?(?P<id>[^/?#&]+)'
_TESTS = [{
'url': 'http://embed.nexx.cloud/748/KC1614647Z27Y7T?autoplay=1',
'md5': '16746bfc28c42049492385c989b26c4a',
'info_dict': {
@ -420,7 +420,6 @@ class NexxEmbedIE(InfoExtractor):
'title': 'Nervenkitzel Achterbahn',
'alt_title': 'Karussellbauer in Deutschland',
'description': 'md5:ffe7b1cc59a01f585e0569949aef73cc',
'release_year': 2005,
'creator': 'SPIEGEL TV',
'thumbnail': r're:^https?://.*\.jpg$',
'duration': 2761,
@ -431,7 +430,10 @@ class NexxEmbedIE(InfoExtractor):
'format': 'bestvideo',
'skip_download': True,
},
}
}, {
'url': 'https://embed.nexx.cloud/11888/video/DSRTO7UVOX06S7',
'only_matching': True,
}]
@staticmethod
def _extract_urls(webpage):

80
youtube_dl/extractor/periscope.py
View File

@ -17,12 +17,54 @@ class PeriscopeBaseIE(InfoExtractor):
'https://api.periscope.tv/api/v2/%s' % method,
item_id, query=query)
def _parse_broadcast_data(self, broadcast, video_id):
title = broadcast['status']
uploader = broadcast.get('user_display_name') or broadcast.get('username')
title = '%s - %s' % (uploader, title) if uploader else title
is_live = broadcast.get('state').lower() == 'running'
thumbnails = [{
'url': broadcast[image],
} for image in ('image_url', 'image_url_small') if broadcast.get(image)]
return {
'id': broadcast.get('id') or video_id,
'title': self._live_title(title) if is_live else title,
'timestamp': parse_iso8601(broadcast.get('created_at')),
'uploader': uploader,
'uploader_id': broadcast.get('user_id') or broadcast.get('username'),
'thumbnails': thumbnails,
'view_count': int_or_none(broadcast.get('total_watched')),
'tags': broadcast.get('tags'),
'is_live': is_live,
}
@staticmethod
def _extract_common_format_info(broadcast):
return broadcast.get('state').lower(), int_or_none(broadcast.get('width')), int_or_none(broadcast.get('height'))
@staticmethod
def _add_width_and_height(f, width, height):
for key, val in (('width', width), ('height', height)):
if not f.get(key):
f[key] = val
def _extract_pscp_m3u8_formats(self, m3u8_url, video_id, format_id, state, width, height, fatal=True):
m3u8_formats = self._extract_m3u8_formats(
m3u8_url, video_id, 'mp4',
entry_protocol='m3u8_native'
if state in ('ended', 'timed_out') else 'm3u8',
m3u8_id=format_id, fatal=fatal)
if len(m3u8_formats) == 1:
self._add_width_and_height(m3u8_formats[0], width, height)
return m3u8_formats
class PeriscopeIE(PeriscopeBaseIE):
IE_DESC = 'Periscope'
IE_NAME = 'periscope'
_VALID_URL = r'https?://(?:www\.)?(?:periscope|pscp)\.tv/[^/]+/(?P<id>[^/?#]+)'
# Alive example URLs can be found here http://onperiscope.com/
# Alive example URLs can be found here https://www.periscope.tv/
_TESTS = [{
'url': 'https://www.periscope.tv/w/aJUQnjY3MjA3ODF8NTYxMDIyMDl2zCg2pECBgwTqRpQuQD352EMPTKQjT4uqlM3cgWFA-g==',
'md5': '65b57957972e503fcbbaeed8f4fa04ca',
@ -61,21 +103,9 @@ class PeriscopeIE(PeriscopeBaseIE):
'accessVideoPublic', {'broadcast_id': token}, token)
broadcast = stream['broadcast']
title = broadcast['status']
info = self._parse_broadcast_data(broadcast, token)
uploader = broadcast.get('user_display_name') or broadcast.get('username')
uploader_id = (broadcast.get('user_id') or broadcast.get('username'))
title = '%s - %s' % (uploader, title) if uploader else title
state = broadcast.get('state').lower()
if state == 'running':
title = self._live_title(title)
timestamp = parse_iso8601(broadcast.get('created_at'))
thumbnails = [{
'url': broadcast[image],
} for image in ('image_url', 'image_url_small') if broadcast.get(image)]
width = int_or_none(broadcast.get('width'))
height = int_or_none(broadcast.get('height'))
@ -92,32 +122,20 @@ class PeriscopeIE(PeriscopeBaseIE):
continue
video_urls.add(video_url)
if format_id != 'rtmp':
m3u8_formats = self._extract_m3u8_formats(
video_url, token, 'mp4',
entry_protocol='m3u8_native'
if state in ('ended', 'timed_out') else 'm3u8',
m3u8_id=format_id, fatal=False)
if len(m3u8_formats) == 1:
add_width_and_height(m3u8_formats[0])
m3u8_formats = self._extract_pscp_m3u8_formats(
video_url, token, format_id, state, width, height, False)
formats.extend(m3u8_formats)
continue
rtmp_format = {
'url': video_url,
'ext': 'flv' if format_id == 'rtmp' else 'mp4',
}
add_width_and_height(rtmp_format)
self._add_width_and_height(rtmp_format)
formats.append(rtmp_format)
self._sort_formats(formats)
return {
'id': broadcast.get('id') or token,
'title': title,
'timestamp': timestamp,
'uploader': uploader,
'uploader_id': uploader_id,
'thumbnails': thumbnails,
'formats': formats,
}
info['formats'] = formats
return info
class PeriscopeUserIE(PeriscopeBaseIE):

170
youtube_dl/extractor/revision3.py
View File

@ -1,170 +0,0 @@
# coding: utf-8
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
int_or_none,
parse_iso8601,
unescapeHTML,
qualities,
)
class Revision3EmbedIE(InfoExtractor):
IE_NAME = 'revision3:embed'
_VALID_URL = r'(?:revision3:(?:(?P<playlist_type>[^:]+):)?|https?://(?:(?:(?:www|embed)\.)?(?:revision3|animalist)|(?:(?:api|embed)\.)?seekernetwork)\.com/player/embed\?videoId=)(?P<playlist_id>\d+)'
_TEST = {
'url': 'http://api.seekernetwork.com/player/embed?videoId=67558',
'md5': '83bcd157cab89ad7318dd7b8c9cf1306',
'info_dict': {
'id': '67558',
'ext': 'mp4',
'title': 'The Pros & Cons Of Zoos',
'description': 'Zoos are often depicted as a terrible place for animals to live, but is there any truth to this?',
'uploader_id': 'dnews',
'uploader': 'DNews',
}
}
_API_KEY = 'ba9c741bce1b9d8e3defcc22193f3651b8867e62'
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
playlist_id = mobj.group('playlist_id')
playlist_type = mobj.group('playlist_type') or 'video_id'
video_data = self._download_json(
'http://revision3.com/api/getPlaylist.json', playlist_id, query={
'api_key': self._API_KEY,
'codecs': 'h264,vp8,theora',
playlist_type: playlist_id,
})['items'][0]
formats = []
for vcodec, media in video_data['media'].items():
for quality_id, quality in media.items():
if quality_id == 'hls':
formats.extend(self._extract_m3u8_formats(
quality['url'], playlist_id, 'mp4',
'm3u8_native', m3u8_id='hls', fatal=False))
else:
formats.append({
'url': quality['url'],
'format_id': '%s-%s' % (vcodec, quality_id),
'tbr': int_or_none(quality.get('bitrate')),
'vcodec': vcodec,
})
self._sort_formats(formats)
return {
'id': playlist_id,
'title': unescapeHTML(video_data['title']),
'description': unescapeHTML(video_data.get('summary')),
'uploader': video_data.get('show', {}).get('name'),
'uploader_id': video_data.get('show', {}).get('slug'),
'duration': int_or_none(video_data.get('duration')),
'formats': formats,
}
class Revision3IE(InfoExtractor):
IE_NAME = 'revision'
_VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:revision3|animalist)\.com)/(?P<id>[^/]+(?:/[^/?#]+)?)'
_TESTS = [{
'url': 'http://www.revision3.com/technobuffalo/5-google-predictions-for-2016',
'md5': 'd94a72d85d0a829766de4deb8daaf7df',
'info_dict': {
'id': '71089',
'display_id': 'technobuffalo/5-google-predictions-for-2016',
'ext': 'webm',
'title': '5 Google Predictions for 2016',
'description': 'Google had a great 2015, but it\'s already time to look ahead. Here are our five predictions for 2016.',
'upload_date': '20151228',
'timestamp': 1451325600,
'duration': 187,
'uploader': 'TechnoBuffalo',
'uploader_id': 'technobuffalo',
}
}, {
# Show
'url': 'http://revision3.com/variant',
'only_matching': True,
}, {
# Tag
'url': 'http://revision3.com/vr',
'only_matching': True,
}]
_PAGE_DATA_TEMPLATE = 'http://www.%s/apiProxy/ddn/%s?domain=%s'
def _real_extract(self, url):
domain, display_id = re.match(self._VALID_URL, url).groups()
site = domain.split('.')[0]
page_info = self._download_json(
self._PAGE_DATA_TEMPLATE % (domain, display_id, domain), display_id)
page_data = page_info['data']
page_type = page_data['type']
if page_type in ('episode', 'embed'):
show_data = page_data['show']['data']
page_id = compat_str(page_data['id'])
video_id = compat_str(page_data['video']['data']['id'])
preference = qualities(['mini', 'small', 'medium', 'large'])
thumbnails = [{
'url': image_url,
'id': image_id,
'preference': preference(image_id)
} for image_id, image_url in page_data.get('images', {}).items()]
info = {
'id': page_id,
'display_id': display_id,
'title': unescapeHTML(page_data['name']),
'description': unescapeHTML(page_data.get('summary')),
'timestamp': parse_iso8601(page_data.get('publishTime'), ' '),
'author': page_data.get('author'),
'uploader': show_data.get('name'),
'uploader_id': show_data.get('slug'),
'thumbnails': thumbnails,
'extractor_key': site,
}
if page_type == 'embed':
info.update({
'_type': 'url_transparent',
'url': page_data['video']['data']['embed'],
})
return info
info.update({
'_type': 'url_transparent',
'url': 'revision3:%s' % video_id,
})
return info
else:
list_data = page_info[page_type]['data']
episodes_data = page_info['episodes']['data']
num_episodes = page_info['meta']['totalEpisodes']
processed_episodes = 0
entries = []
page_num = 1
while True:
entries.extend([{
'_type': 'url',
'url': 'http://%s%s' % (domain, episode['path']),
'id': compat_str(episode['id']),
'ie_key': 'Revision3',
'extractor_key': site,
} for episode in episodes_data])
processed_episodes += len(episodes_data)
if processed_episodes == num_episodes:
break
page_num += 1
episodes_data = self._download_json(self._PAGE_DATA_TEMPLATE % (
domain, display_id + '/' + compat_str(page_num), domain),
display_id)['episodes']['data']
return self.playlist_result(
entries, compat_str(list_data['id']),
list_data.get('name'), list_data.get('summary'))

45
youtube_dl/extractor/seeker.py
View File

@ -4,34 +4,37 @@ from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..utils import (
get_element_by_class,
strip_or_none,
)
class SeekerIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?seeker\.com/(?P<display_id>.*)-(?P<article_id>\d+)\.html'
_TESTS = [{
# player.loadRevision3Item
'url': 'http://www.seeker.com/should-trump-be-required-to-release-his-tax-returns-1833805621.html',
'md5': '30c1dc4030cc715cf05b423d0947ac18',
'md5': '897d44bbe0d8986a2ead96de565a92db',
'info_dict': {
'id': '76243',
'ext': 'webm',
'id': 'Elrn3gnY',
'ext': 'mp4',
'title': 'Should Trump Be Required To Release His Tax Returns?',
'description': 'Donald Trump has been secretive about his "big," "beautiful" tax returns. So what can we learn if he decides to release them?',
'uploader': 'Seeker Daily',
'uploader_id': 'seekerdaily',
'description': 'md5:41efa8cfa8d627841045eec7b018eb45',
'timestamp': 1490090165,
'upload_date': '20170321',
}
}, {
'url': 'http://www.seeker.com/changes-expected-at-zoos-following-recent-gorilla-lion-shootings-1834116536.html',
'playlist': [
{
'md5': '83bcd157cab89ad7318dd7b8c9cf1306',
'md5': '0497b9f20495174be73ae136949707d2',
'info_dict': {
'id': '67558',
'id': 'FihYQ8AE',
'ext': 'mp4',
'title': 'The Pros & Cons Of Zoos',
'description': 'Zoos are often depicted as a terrible place for animals to live, but is there any truth to this?',
'uploader': 'DNews',
'uploader_id': 'dnews',
'description': 'md5:d88f99a8ea8e7d25e6ff77f271b1271c',
'timestamp': 1490039133,
'upload_date': '20170320',
},
}
],
@ -45,13 +48,11 @@ class SeekerIE(InfoExtractor):
def _real_extract(self, url):
display_id, article_id = re.match(self._VALID_URL, url).groups()
webpage = self._download_webpage(url, display_id)
mobj = re.search(r"player\.loadRevision3Item\('([^']+)'\s*,\s*(\d+)\);", webpage)
if mobj:
playlist_type, playlist_id = mobj.groups()
return self.url_result(
'revision3:%s:%s' % (playlist_type, playlist_id), 'Revision3Embed', playlist_id)
else:
entries = [self.url_result('revision3:video_id:%s' % video_id, 'Revision3Embed', video_id) for video_id in re.findall(
r'<iframe[^>]+src=[\'"](?:https?:)?//api\.seekernetwork\.com/player/embed\?videoId=(\d+)', webpage)]
return self.playlist_result(
entries, article_id, self._og_search_title(webpage), self._og_search_description(webpage))
entries = []
for jwp_id in re.findall(r'data-video-id="([a-zA-Z0-9]{8})"', webpage):
entries.append(self.url_result(
'jwplatform:' + jwp_id, 'JWPlatform', jwp_id))
return self.playlist_result(
entries, article_id,
self._og_search_title(webpage),
strip_or_none(get_element_by_class('subtitle__text', webpage)) or self._og_search_description(webpage))

2
youtube_dl/extractor/soundcloud.py
View File

@ -276,7 +276,7 @@ class SoundcloudIE(InfoExtractor):
if secret_token:
query['secret_token'] = secret_token
if info.get('downloadable'):
if info.get('downloadable') and info.get('has_downloads_left'):
format_url = update_url_query(
info.get('download_url') or track_base_url + '/download', query)
format_urls.add(format_url)

68
youtube_dl/extractor/teamcoco.py
View File

@ -84,6 +84,19 @@ class TeamcocoIE(TurnerBaseIE):
'only_matching': True,
}
]
_RECORD_TEMPL = '''id
title
teaser
publishOn
thumb {
preview
}
tags {
name
}
duration
turnerMediaId
turnerMediaAuthToken'''
def _graphql_call(self, query_template, object_type, object_id):
find_object = 'find' + object_type
@ -98,36 +111,36 @@ class TeamcocoIE(TurnerBaseIE):
display_id = self._match_id(url)
response = self._graphql_call('''{
%s(slug: "%s") {
%%s(slug: "%%s") {
... on RecordSlug {
record {
%s
}
}
... on PageSlug {
child {
id
title
teaser
publishOn
thumb {
preview
}
file {
url
}
tags {
name
}
duration
turnerMediaId
turnerMediaAuthToken
}
}
... on NotFoundSlug {
status
}
}
}''', 'Slug', display_id)
}''' % self._RECORD_TEMPL, 'Slug', display_id)
if response.get('status'):
raise ExtractorError('This video is no longer available.', expected=True)
record = response['record']
child = response.get('child')
if child:
record = self._graphql_call('''{
%%s(id: "%%s") {
... on Video {
%s
}
}
}''' % self._RECORD_TEMPL, 'Record', child['id'])
else:
record = response['record']
video_id = record['id']
info = {
@ -150,25 +163,21 @@ class TeamcocoIE(TurnerBaseIE):
'accessTokenType': 'jws',
}))
else:
d = self._download_json(
video_sources = self._download_json(
'https://teamcoco.com/_truman/d/' + video_id,
video_id, fatal=False) or {}
video_sources = d.get('meta') or {}
if not video_sources:
video_sources = self._graphql_call('''{
%s(id: "%s") {
src
}
}''', 'RecordVideoSource', video_id) or {}
video_id)['meta']['src']
if isinstance(video_sources, dict):
video_sources = video_sources.values()
formats = []
get_quality = qualities(['low', 'sd', 'hd', 'uhd'])
for format_id, src in video_sources.get('src', {}).items():
for src in video_sources:
if not isinstance(src, dict):
continue
src_url = src.get('src')
if not src_url:
continue
format_id = src.get('label')
ext = determine_ext(src_url, mimetype2ext(src.get('type')))
if format_id == 'hls' or ext == 'm3u8':
# compat_urllib_parse.urljoin does not work here
@ -190,9 +199,6 @@ class TeamcocoIE(TurnerBaseIE):
'format_id': format_id,
'quality': get_quality(format_id),
})
if not formats:
formats = self._extract_m3u8_formats(
record['file']['url'], video_id, 'mp4', fatal=False)
self._sort_formats(formats)
info['formats'] = formats

10
youtube_dl/extractor/twitch.py
View File

@ -344,9 +344,8 @@ class TwitchVodIE(TwitchItemBaseIE):
info['subtitles'] = {
'rechat': [{
'url': update_url_query(
'https://rechat.twitch.tv/rechat-messages', {
'video_id': 'v%s' % item_id,
'start': info['timestamp'],
'https://api.twitch.tv/v5/videos/%s/comments' % item_id, {
'client_id': self._CLIENT_ID,
}),
'ext': 'json',
}],
@ -644,7 +643,7 @@ class TwitchStreamIE(TwitchBaseIE):
class TwitchClipsIE(TwitchBaseIE):
IE_NAME = 'twitch:clips'
_VALID_URL = r'https?://(?:clips\.twitch\.tv/(?:[^/]+/)*|(?:www\.)?twitch\.tv/[^/]+/clip/)(?P<id>[^/?#&]+)'
_VALID_URL = r'https?://(?:clips\.twitch\.tv/(?:embed\?.*?\bclip=|(?:[^/]+/)*)|(?:www\.)?twitch\.tv/[^/]+/clip/)(?P<id>[^/?#&]+)'
_TESTS = [{
'url': 'https://clips.twitch.tv/FaintLightGullWholeWheat',
@ -667,6 +666,9 @@ class TwitchClipsIE(TwitchBaseIE):
}, {
'url': 'https://www.twitch.tv/sergeynixon/clip/StormyThankfulSproutFutureMan',
'only_matching': True,
}, {
'url': 'https://clips.twitch.tv/embed?clip=InquisitiveBreakableYogurtJebaited',
'only_matching': True,
}]
def _real_extract(self, url):

582
youtube_dl/extractor/twitter.py
View File

@ -2,35 +2,69 @@
from __future__ import unicode_literals
import re
import time
from .common import InfoExtractor
from ..compat import compat_urlparse
from ..compat import (
compat_HTTPError,
compat_parse_qs,
compat_urllib_parse_unquote,
compat_urllib_parse_urlparse,
)
from ..utils import (
determine_ext,
dict_get,
ExtractorError,
float_or_none,
int_or_none,
remove_end,
try_get,
strip_or_none,
unified_timestamp,
update_url_query,
xpath_text,
parse_count
)
from .periscope import PeriscopeIE
from .periscope import (
PeriscopeBaseIE,
PeriscopeIE,
)
class TwitterBaseIE(InfoExtractor):
_API_BASE = 'https://api.twitter.com/1.1/'
_BASE_REGEX = r'https?://(?:(?:www|m(?:obile)?)\.)?twitter\.com/'
_GUEST_TOKEN = None
def _extract_variant_formats(self, variant, video_id):
variant_url = variant.get('url')
if not variant_url:
return []
elif '.m3u8' in variant_url:
return self._extract_m3u8_formats(
variant_url, video_id, 'mp4', 'm3u8_native',
m3u8_id='hls', fatal=False)
else:
tbr = int_or_none(dict_get(variant, ('bitrate', 'bit_rate')), 1000) or None
f = {
'url': variant_url,
'format_id': 'http' + ('-%d' % tbr if tbr else ''),
'tbr': tbr,
}
self._search_dimensions_in_video_url(f, variant_url)
return [f]
def _extract_formats_from_vmap_url(self, vmap_url, video_id):
vmap_data = self._download_xml(vmap_url, video_id)
video_url = xpath_text(vmap_data, './/MediaFile').strip()
if determine_ext(video_url) == 'm3u8':
return self._extract_m3u8_formats(
video_url, video_id, ext='mp4', m3u8_id='hls',
entry_protocol='m3u8_native')
return [{
'url': video_url,
}]
formats = []
urls = []
for video_variant in vmap_data.findall('.//{http://twitter.com/schema/videoVMapV2.xsd}videoVariant'):
video_variant.attrib['url'] = compat_urllib_parse_unquote(
video_variant.attrib['url'])
urls.append(video_variant.attrib['url'])
formats.extend(self._extract_variant_formats(
video_variant.attrib, video_id))
video_url = strip_or_none(xpath_text(vmap_data, './/MediaFile'))
if video_url not in urls:
formats.extend(self._extract_variant_formats({'url': video_url}, video_id))
return formats
@staticmethod
def _search_dimensions_in_video_url(a_format, video_url):
@ -41,10 +75,30 @@ class TwitterBaseIE(InfoExtractor):
'height': int(m.group('height')),
})
def _call_api(self, path, video_id, query={}):
headers = {
'Authorization': 'Bearer AAAAAAAAAAAAAAAAAAAAAPYXBAAAAAAACLXUNDekMxqa8h%2F40K4moUkGsoc%3DTYfbDKbT3jJPCEVnMYqilB28NHfOPqkca3qaAxGfsyKCs0wRbw',
}
if not self._GUEST_TOKEN:
self._GUEST_TOKEN = self._download_json(
self._API_BASE + 'guest/activate.json', video_id,
'Downloading guest token', data=b'',
headers=headers)['guest_token']
headers['x-guest-token'] = self._GUEST_TOKEN
try:
return self._download_json(
self._API_BASE + path, video_id, headers=headers, query=query)
except ExtractorError as e:
if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
raise ExtractorError(self._parse_json(
e.cause.read().decode(),
video_id)['errors'][0]['message'], expected=True)
raise
class TwitterCardIE(TwitterBaseIE):
class TwitterCardIE(InfoExtractor):
IE_NAME = 'twitter:card'
_VALID_URL = r'https?://(?:www\.)?twitter\.com/i/(?P<path>cards/tfw/v1|videos(?:/tweet)?)/(?P<id>\d+)'
_VALID_URL = TwitterBaseIE._BASE_REGEX + r'i/(?:cards/tfw/v1|videos(?:/tweet)?)/(?P<id>\d+)'
_TESTS = [
{
'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
@ -52,19 +106,28 @@ class TwitterCardIE(TwitterBaseIE):
'info_dict': {
'id': '560070183650213889',
'ext': 'mp4',
'title': 'Twitter web player',
'thumbnail': r're:^https?://.*\.jpg$',
'title': "Twitter - You can now shoot, edit and share video on Twitter. Capture life's most moving moments from your perspective.",
'description': 'md5:18d3e24bb4f6e5007487dd546e53bd96',
'uploader': 'Twitter',
'uploader_id': 'Twitter',
'thumbnail': r're:^https?://.*\.jpg',
'duration': 30.033,
'timestamp': 1422366112,
'upload_date': '20150127',
},
},
{
'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768',
'md5': '7ee2a553b63d1bccba97fbed97d9e1c8',
'md5': '7137eca597f72b9abbe61e5ae0161399',
'info_dict': {
'id': '623160978427936768',
'ext': 'mp4',
'title': 'Twitter web player',
'thumbnail': r're:^https?://.*$',
'title': "NASA - Fly over Pluto's icy Norgay Mountains and Sputnik Plain in this @NASANewHorizons #PlutoFlyby video.",
'description': "Fly over Pluto's icy Norgay Mountains and Sputnik Plain in this @NASANewHorizons #PlutoFlyby video. https://t.co/BJYgOjSeGA",
'uploader': 'NASA',
'uploader_id': 'NASA',
'timestamp': 1437408129,
'upload_date': '20150720',
},
},
{
@ -76,7 +139,7 @@ class TwitterCardIE(TwitterBaseIE):
'title': 'Ubuntu 11.10 Overview',
'description': 'md5:a831e97fa384863d6e26ce48d1c43376',
'upload_date': '20111013',
'uploader': 'OMG! Ubuntu!',
'uploader': 'OMG! UBUNTU!',
'uploader_id': 'omgubuntu',
},
'add_ie': ['Youtube'],
@ -100,199 +163,30 @@ class TwitterCardIE(TwitterBaseIE):
'info_dict': {
'id': '705235433198714880',
'ext': 'mp4',
'title': 'Twitter web player',
'thumbnail': r're:^https?://.*',
'title': "Brent Yarina - Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight.",
'description': "Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight. https://t.co/OrxcJ28Bns",
'uploader': 'Brent Yarina',
'uploader_id': 'BTNBrentYarina',
'timestamp': 1456976204,
'upload_date': '20160303',
},
'skip': 'This content is no longer available.',
}, {
'url': 'https://twitter.com/i/videos/752274308186120192',
'only_matching': True,
},
]
_API_BASE = 'https://api.twitter.com/1.1'
def _parse_media_info(self, media_info, video_id):
formats = []
for media_variant in media_info.get('variants', []):
media_url = media_variant['url']
if media_url.endswith('.m3u8'):
formats.extend(self._extract_m3u8_formats(media_url, video_id, ext='mp4', m3u8_id='hls'))
elif media_url.endswith('.mpd'):
formats.extend(self._extract_mpd_formats(media_url, video_id, mpd_id='dash'))
else:
tbr = int_or_none(dict_get(media_variant, ('bitRate', 'bitrate')), scale=1000)
a_format = {
'url': media_url,
'format_id': 'http-%d' % tbr if tbr else 'http',
'tbr': tbr,
}
# Reported bitRate may be zero
if not a_format['tbr']:
del a_format['tbr']
self._search_dimensions_in_video_url(a_format, media_url)
formats.append(a_format)
return formats
def _extract_mobile_formats(self, username, video_id):
webpage = self._download_webpage(
'https://mobile.twitter.com/%s/status/%s' % (username, video_id),
video_id, 'Downloading mobile webpage',
headers={
# A recent mobile UA is necessary for `gt` cookie
'User-Agent': 'Mozilla/5.0 (Android 6.0.1; Mobile; rv:54.0) Gecko/54.0 Firefox/54.0',
})
main_script_url = self._html_search_regex(
r'<script[^>]+src="([^"]+main\.[^"]+)"', webpage, 'main script URL')
main_script = self._download_webpage(
main_script_url, video_id, 'Downloading main script')
bearer_token = self._search_regex(
r'BEARER_TOKEN\s*:\s*"([^"]+)"',
main_script, 'bearer token')
# https://developer.twitter.com/en/docs/tweets/post-and-engage/api-reference/get-statuses-show-id
api_data = self._download_json(
'%s/statuses/show/%s.json' % (self._API_BASE, video_id),
video_id, 'Downloading API data',
headers={
'Authorization': 'Bearer ' + bearer_token,
})
media_info = try_get(api_data, lambda o: o['extended_entities']['media'][0]['video_info']) or {}
return self._parse_media_info(media_info, video_id)
def _real_extract(self, url):
path, video_id = re.search(self._VALID_URL, url).groups()
config = None
formats = []
duration = None
view_count = None
urls = [url]
if path.startswith('cards/'):
urls.append('https://twitter.com/i/videos/' + video_id)
for u in urls:
webpage = self._download_webpage(
u, video_id, headers={'Referer': 'https://twitter.com/'})
iframe_url = self._html_search_regex(
r'<iframe[^>]+src="((?:https?:)?//(?:www\.youtube\.com/embed/[^"]+|(?:www\.)?vine\.co/v/\w+/card))"',
webpage, 'video iframe', default=None)
if iframe_url:
return self.url_result(iframe_url)
config = self._parse_json(self._html_search_regex(
r'data-(?:player-)?config="([^"]+)"', webpage,
'data player config', default='{}'),
video_id)
if config.get('source_type') == 'vine':
return self.url_result(config['player_url'], 'Vine')
periscope_url = PeriscopeIE._extract_url(webpage)
if periscope_url:
return self.url_result(periscope_url, PeriscopeIE.ie_key())
video_url = config.get('video_url') or config.get('playlist', [{}])[0].get('source')
if video_url:
if determine_ext(video_url) == 'm3u8':
formats.extend(self._extract_m3u8_formats(video_url, video_id, ext='mp4', m3u8_id='hls'))
else:
f = {
'url': video_url,
}
self._search_dimensions_in_video_url(f, video_url)
formats.append(f)
vmap_url = config.get('vmapUrl') or config.get('vmap_url')
if vmap_url:
formats.extend(
self._extract_formats_from_vmap_url(vmap_url, video_id))
media_info = None
for entity in config.get('status', {}).get('entities', []):
if 'mediaInfo' in entity:
media_info = entity['mediaInfo']
if media_info:
formats.extend(self._parse_media_info(media_info, video_id))
duration = float_or_none(media_info.get('duration', {}).get('nanos'), scale=1e9)
username = config.get('user', {}).get('screen_name')
if username:
formats.extend(self._extract_mobile_formats(username, video_id))
if formats:
title = self._search_regex(r'<title>([^<]+)</title>', webpage, 'title')
thumbnail = config.get('posterImageUrl') or config.get('image_src')
duration = float_or_none(config.get('duration'), scale=1000) or duration
break
if not formats:
headers = {
'Authorization': 'Bearer AAAAAAAAAAAAAAAAAAAAAPYXBAAAAAAACLXUNDekMxqa8h%2F40K4moUkGsoc%3DTYfbDKbT3jJPCEVnMYqilB28NHfOPqkca3qaAxGfsyKCs0wRbw',
'Referer': url,
}
ct0 = self._get_cookies(url).get('ct0')
if ct0:
headers['csrf_token'] = ct0.value
guest_token_c = self._get_cookies('http://api.twitter.com/').get('gt')
if not guest_token_c:
guest_token = self._download_json(
'%s/guest/activate.json' % self._API_BASE, video_id,
'Downloading guest token', data=b'',
headers=headers)['guest_token']
self._set_cookie('api.twitter.com', 'gt', guest_token, expire_time=time.time() + 3000)
else:
guest_token = guest_token_c.value
headers['x-guest-token'] = guest_token
config = self._download_json(
'%s/videos/tweet/config/%s.json' % (self._API_BASE, video_id),
video_id, headers=headers)
track = config['track']
vmap_url = track.get('vmapUrl')
if vmap_url:
formats = self._extract_formats_from_vmap_url(vmap_url, video_id)
else:
playback_url = track['playbackUrl']
if determine_ext(playback_url) == 'm3u8':
formats = self._extract_m3u8_formats(
playback_url, video_id, 'mp4',
entry_protocol='m3u8_native', m3u8_id='hls')
else:
formats = [{
'url': playback_url,
}]
title = 'Twitter web player'
thumbnail = config.get('posterImage')
duration = float_or_none(track.get('durationMs'), scale=1000)
view_count = parse_count(track.get('viewCount'))
self._remove_duplicate_formats(formats)
self._sort_formats(formats)
return {
'id': video_id,
'title': title,
'thumbnail': thumbnail,
'duration': duration,
'formats': formats,
'view_count': view_count
}
status_id = self._match_id(url)
return self.url_result(
'https://twitter.com/statuses/' + status_id,
TwitterIE.ie_key(), status_id)
class TwitterIE(InfoExtractor):
class TwitterIE(TwitterBaseIE):
IE_NAME = 'twitter'
_VALID_URL = r'https?://(?:www\.|m\.|mobile\.)?twitter\.com/(?:i/web|(?P<user_id>[^/]+))/status/(?P<id>\d+)'
_TEMPLATE_URL = 'https://twitter.com/%s/status/%s'
_TEMPLATE_STATUSES_URL = 'https://twitter.com/statuses/%s'
_VALID_URL = TwitterBaseIE._BASE_REGEX + r'(?:(?:i/web|[^/]+)/status|statuses)/(?P<id>\d+)'
_TESTS = [{
'url': 'https://twitter.com/freethenipple/status/643211948184596480',
@ -301,10 +195,13 @@ class TwitterIE(InfoExtractor):
'ext': 'mp4',
'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!',
'thumbnail': r're:^https?://.*\.jpg',
'description': 'FREE THE NIPPLE on Twitter: "FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ"',
'description': 'FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ',
'uploader': 'FREE THE NIPPLE',
'uploader_id': 'freethenipple',
'duration': 12.922,
'timestamp': 1442188653,
'upload_date': '20150913',
'age_limit': 18,
},
}, {
'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1',
@ -326,19 +223,23 @@ class TwitterIE(InfoExtractor):
'id': '665052190608723968',
'ext': 'mp4',
'title': 'Star Wars - A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens.',
'description': 'Star Wars on Twitter: "A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens."',
'description': 'A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens. https://t.co/OkSqT2fjWJ',
'uploader_id': 'starwars',
'uploader': 'Star Wars',
'timestamp': 1447395772,
'upload_date': '20151113',
},
}, {
'url': 'https://twitter.com/BTNBrentYarina/status/705235433198714880',
'info_dict': {
'id': '705235433198714880',
'ext': 'mp4',
'title': 'Brent Yarina - Khalil Iverson\'s missed highlight dunk. And made highlight dunk. In one highlight.',
'description': 'Brent Yarina on Twitter: "Khalil Iverson\'s missed highlight dunk. And made highlight dunk. In one highlight."',
'title': "Brent Yarina - Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight.",
'description': "Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight. https://t.co/OrxcJ28Bns",
'uploader_id': 'BTNBrentYarina',
'uploader': 'Brent Yarina',
'timestamp': 1456976204,
'upload_date': '20160303',
},
'params': {
# The same video as https://twitter.com/i/videos/tweet/705235433198714880
@ -350,12 +251,14 @@ class TwitterIE(InfoExtractor):
'info_dict': {
'id': '700207533655363584',
'ext': 'mp4',
'title': 'JG - BEAT PROD: @suhmeduh #Damndaniel',
'description': 'JG on Twitter: "BEAT PROD: @suhmeduh https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ"',
'title': 'Simon Vertugo - BEAT PROD: @suhmeduh #Damndaniel',
'description': 'BEAT PROD: @suhmeduh https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ',
'thumbnail': r're:^https?://.*\.jpg',
'uploader': 'JG',
'uploader_id': 'jaydingeer',
'uploader': 'Simon Vertugo',
'uploader_id': 'simonvertugo',
'duration': 30.0,
'timestamp': 1455777459,
'upload_date': '20160218',
},
}, {
'url': 'https://twitter.com/Filmdrunk/status/713801302971588609',
@ -363,10 +266,9 @@ class TwitterIE(InfoExtractor):
'info_dict': {
'id': 'MIOxnrUteUd',
'ext': 'mp4',
'title': 'Vince Mancini - Vine of the day',
'description': 'Vince Mancini on Twitter: "Vine of the day https://t.co/xmTvRdqxWf"',
'uploader': 'Vince Mancini',
'uploader_id': 'Filmdrunk',
'title': 'Dr.Pepperの飲み方 #japanese #バカ #ドクペ #電動ガン',
'uploader': 'TAKUMA',
'uploader_id': '1004126642786242560',
'timestamp': 1402826626,
'upload_date': '20140615',
},
@ -377,21 +279,22 @@ class TwitterIE(InfoExtractor):
'id': '719944021058060289',
'ext': 'mp4',
'title': 'Captain America - @King0fNerd Are you sure you made the right choice? Find out in theaters.',
'description': 'Captain America on Twitter: "@King0fNerd Are you sure you made the right choice? Find out in theaters. https://t.co/GpgYi9xMJI"',
'uploader_id': 'captainamerica',
'description': '@King0fNerd Are you sure you made the right choice? Find out in theaters. https://t.co/GpgYi9xMJI',
'uploader_id': 'CaptainAmerica',
'uploader': 'Captain America',
'duration': 3.17,
'timestamp': 1460483005,
'upload_date': '20160412',
},
}, {
'url': 'https://twitter.com/OPP_HSD/status/779210622571536384',
'info_dict': {
'id': '1zqKVVlkqLaKB',
'ext': 'mp4',
'title': 'Sgt Kerry Schmidt - LIVE on #Periscope: Road rage, mischief, assault, rollover and fire in one occurrence',
'description': 'Sgt Kerry Schmidt on Twitter: "LIVE on #Periscope: Road rage, mischief, assault, rollover and fire in one occurrence https://t.co/EKrVgIXF3s"',
'title': 'Sgt Kerry Schmidt - Ontario Provincial Police - Road rage, mischief, assault, rollover and fire in one occurrence',
'upload_date': '20160923',
'uploader_id': 'OPP_HSD',
'uploader': 'Sgt Kerry Schmidt',
'uploader_id': '1PmKqpJdOJQoY',
'uploader': 'Sgt Kerry Schmidt - Ontario Provincial Police',
'timestamp': 1474613214,
},
'add_ie': ['Periscope'],
@ -402,10 +305,12 @@ class TwitterIE(InfoExtractor):
'id': '852138619213144067',
'ext': 'mp4',
'title': 'عالم الأخبار - كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة',
'description': 'عالم الأخبار on Twitter: "كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة https://t.co/xg6OhpyKfN"',
'description': 'كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة https://t.co/xg6OhpyKfN',
'uploader': 'عالم الأخبار',
'uploader_id': 'news_al3alm',
'duration': 277.4,
'timestamp': 1492000653,
'upload_date': '20170412',
},
}, {
'url': 'https://twitter.com/i/web/status/910031516746514432',
@ -414,10 +319,12 @@ class TwitterIE(InfoExtractor):
'ext': 'mp4',
'title': 'Préfet de Guadeloupe - [Direct] #Maria Le centre se trouve actuellement au sud de Basse-Terre. Restez confinés. Réfugiez-vous dans la pièce la + sûre.',
'thumbnail': r're:^https?://.*\.jpg',
'description': 'Préfet de Guadeloupe on Twitter: "[Direct] #Maria Le centre se trouve actuellement au sud de Basse-Terre. Restez confinés. Réfugiez-vous dans la pièce la + sûre. https://t.co/mwx01Rs4lo"',
'description': '[Direct] #Maria Le centre se trouve actuellement au sud de Basse-Terre. Restez confinés. Réfugiez-vous dans la pièce la + sûre. https://t.co/mwx01Rs4lo',
'uploader': 'Préfet de Guadeloupe',
'uploader_id': 'Prefet971',
'duration': 47.48,
'timestamp': 1505803395,
'upload_date': '20170919',
},
'params': {
'skip_download': True, # requires ffmpeg
@ -430,10 +337,12 @@ class TwitterIE(InfoExtractor):
'ext': 'mp4',
'title': 're:.*?Shep is on a roll today.*?',
'thumbnail': r're:^https?://.*\.jpg',
'description': 'md5:63b036c228772523ae1924d5f8e5ed6b',
'description': 'md5:37b9f2ff31720cef23b2bd42ee8a0f09',
'uploader': 'Lis Power',
'uploader_id': 'LisPower1',
'duration': 111.278,
'timestamp': 1527623489,
'upload_date': '20180529',
},
'params': {
'skip_download': True, # requires ffmpeg
@ -445,88 +354,163 @@ class TwitterIE(InfoExtractor):
'ext': 'mp4',
'title': 'Twitter - A new is coming. Some of you got an opt-in to try it now. Check out the emoji button, quick keyboard shortcuts, upgraded trends, advanced search, and more. Let us know your thoughts!',
'thumbnail': r're:^https?://.*\.jpg',
'description': 'md5:66d493500c013e3e2d434195746a7f78',
'description': 'md5:6dfd341a3310fb97d80d2bf7145df976',
'uploader': 'Twitter',
'uploader_id': 'Twitter',
'duration': 61.567,
'timestamp': 1548184644,
'upload_date': '20190122',
},
}, {
# not available in Periscope
'url': 'https://twitter.com/ViviEducation/status/1136534865145286656',
'info_dict': {
'id': '1vOGwqejwoWxB',
'ext': 'mp4',
'title': 'Vivi - Vivi founder @lior_rauchy announcing our new student feedback tool live at @EduTECH_AU #EduTECH2019',
'uploader': 'Vivi',
'uploader_id': '1eVjYOLGkGrQL',
},
'add_ie': ['TwitterBroadcast'],
}, {
# Twitch Clip Embed
'url': 'https://twitter.com/GunB1g/status/1163218564784017422',
'only_matching': True,
}]
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
twid = mobj.group('id')
webpage, urlh = self._download_webpage_handle(
self._TEMPLATE_STATUSES_URL % twid, twid)
if 'twitter.com/account/suspended' in urlh.geturl():
raise ExtractorError('Account suspended by Twitter.', expected=True)
user_id = None
redirect_mobj = re.match(self._VALID_URL, urlh.geturl())
if redirect_mobj:
user_id = redirect_mobj.group('user_id')
if not user_id:
user_id = mobj.group('user_id')
username = remove_end(self._og_search_title(webpage), ' on Twitter')
title = description = self._og_search_description(webpage).strip('').replace('\n', ' ').strip('“”')
twid = self._match_id(url)
status = self._call_api(
'statuses/show/%s.json' % twid, twid, {
'cards_platform': 'Web-12',
'include_cards': 1,
'include_reply_count': 1,
'include_user_entities': 0,
'tweet_mode': 'extended',
})
title = description = status['full_text'].replace('\n', ' ')
# strip 'https -_t.co_BJYgOjSeGA' junk from filenames
title = re.sub(r'\s+(https?://[^ ]+)', '', title)
user = status.get('user') or {}
uploader = user.get('name')
if uploader:
title = '%s - %s' % (uploader, title)
uploader_id = user.get('screen_name')
tags = []
for hashtag in (try_get(status, lambda x: x['entities']['hashtags'], list) or []):
hashtag_text = hashtag.get('text')
if not hashtag_text:
continue
tags.append(hashtag_text)
info = {
'uploader_id': user_id,
'uploader': username,
'webpage_url': url,
'description': '%s on Twitter: "%s"' % (username, description),
'title': username + ' - ' + title,
'id': twid,
'title': title,
'description': description,
'uploader': uploader,
'timestamp': unified_timestamp(status.get('created_at')),
'uploader_id': uploader_id,
'uploader_url': 'https://twitter.com/' + uploader_id if uploader_id else None,
'like_count': int_or_none(status.get('favorite_count')),
'repost_count': int_or_none(status.get('retweet_count')),
'comment_count': int_or_none(status.get('reply_count')),
'age_limit': 18 if status.get('possibly_sensitive') else 0,
'tags': tags,
}
mobj = re.search(r'''(?x)
<video[^>]+class="animated-gif"(?P<more_info>[^>]+)>\s*
<source[^>]+video-src="(?P<url>[^"]+)"
''', webpage)
media = try_get(status, lambda x: x['extended_entities']['media'][0])
if media and media.get('type') != 'photo':
video_info = media.get('video_info') or {}
formats = []
for variant in video_info.get('variants', []):
formats.extend(self._extract_variant_formats(variant, twid))
self._sort_formats(formats)
thumbnails = []
media_url = media.get('media_url_https') or media.get('media_url')
if media_url:
def add_thumbnail(name, size):
thumbnails.append({
'id': name,
'url': update_url_query(media_url, {'name': name}),
'width': int_or_none(size.get('w') or size.get('width')),
'height': int_or_none(size.get('h') or size.get('height')),
})
for name, size in media.get('sizes', {}).items():
add_thumbnail(name, size)
add_thumbnail('orig', media.get('original_info') or {})
if mobj:
more_info = mobj.group('more_info')
height = int_or_none(self._search_regex(
r'data-height="(\d+)"', more_info, 'height', fatal=False))
width = int_or_none(self._search_regex(
r'data-width="(\d+)"', more_info, 'width', fatal=False))
thumbnail = self._search_regex(
r'poster="([^"]+)"', more_info, 'poster', fatal=False)
info.update({
'id': twid,
'url': mobj.group('url'),
'height': height,
'width': width,
'thumbnail': thumbnail,
'formats': formats,
'thumbnails': thumbnails,
'duration': float_or_none(video_info.get('duration_millis'), 1000),
})
return info
twitter_card_url = None
if 'class="PlayableMedia' in webpage:
twitter_card_url = '%s//twitter.com/i/videos/tweet/%s' % (self.http_scheme(), twid)
else:
twitter_card_iframe_url = self._search_regex(
r'data-full-card-iframe-url=([\'"])(?P<url>(?:(?!\1).)+)\1',
webpage, 'Twitter card iframe URL', default=None, group='url')
if twitter_card_iframe_url:
twitter_card_url = compat_urlparse.urljoin(url, twitter_card_iframe_url)
card = status.get('card')
if card:
binding_values = card['binding_values']
if twitter_card_url:
info.update({
'_type': 'url_transparent',
'ie_key': 'TwitterCard',
'url': twitter_card_url,
})
return info
def get_binding_value(k):
o = binding_values.get(k) or {}
return try_get(o, lambda x: x[x['type'].lower() + '_value'])
raise ExtractorError('There\'s no video in this tweet.')
card_name = card['name'].split(':')[-1]
if card_name == 'amplify':
formats = self._extract_formats_from_vmap_url(
get_binding_value('amplify_url_vmap'),
get_binding_value('amplify_content_id') or twid)
self._sort_formats(formats)
thumbnails = []
for suffix in ('_small', '', '_large', '_x_large', '_original'):
image = get_binding_value('player_image' + suffix) or {}
image_url = image.get('url')
if not image_url or '/player-placeholder' in image_url:
continue
thumbnails.append({
'id': suffix[1:] if suffix else 'medium',
'url': image_url,
'width': int_or_none(image.get('width')),
'height': int_or_none(image.get('height')),
})
info.update({
'formats': formats,
'thumbnails': thumbnails,
'duration': int_or_none(get_binding_value(
'content_duration_seconds')),
})
elif card_name == 'player':
info.update({
'_type': 'url',
'url': get_binding_value('player_url'),
})
elif card_name == 'periscope_broadcast':
info.update({
'_type': 'url',
'url': get_binding_value('url') or get_binding_value('player_url'),
'ie_key': PeriscopeIE.ie_key(),
})
elif card_name == 'broadcast':
info.update({
'_type': 'url',
'url': get_binding_value('broadcast_url'),
'ie_key': TwitterBroadcastIE.ie_key(),
})
else:
raise ExtractorError('Unsupported Twitter Card.')
else:
expanded_url = try_get(status, lambda x: x['entities']['urls'][0]['expanded_url'])
if not expanded_url:
raise ExtractorError("There's no video in this tweet.")
info.update({
'_type': 'url',
'url': expanded_url,
})
return info
class TwitterAmplifyIE(TwitterBaseIE):
@ -583,3 +567,27 @@ class TwitterAmplifyIE(TwitterBaseIE):
'formats': formats,
'thumbnails': thumbnails,
}
class TwitterBroadcastIE(TwitterBaseIE, PeriscopeBaseIE):
IE_NAME = 'twitter:broadcast'
_VALID_URL = TwitterBaseIE._BASE_REGEX + r'i/broadcasts/(?P<id>[0-9a-zA-Z]{13})'
def _real_extract(self, url):
broadcast_id = self._match_id(url)
broadcast = self._call_api(
'broadcasts/show.json', broadcast_id,
{'ids': broadcast_id})['broadcasts'][broadcast_id]
info = self._parse_broadcast_data(broadcast, broadcast_id)
media_key = broadcast['media_key']
source = self._call_api(
'live_video_stream/status/' + media_key, media_key)['source']
m3u8_url = source.get('noRedirectPlaybackUrl') or source['location']
if '/live_video_stream/geoblocked/' in m3u8_url:
self.raise_geo_restricted()
m3u8_id = compat_parse_qs(compat_urllib_parse_urlparse(
m3u8_url).query).get('type', [None])[0]
state, width, height = self._extract_common_format_info(broadcast)
info['formats'] = self._extract_pscp_m3u8_formats(
m3u8_url, broadcast_id, m3u8_id, state, width, height)
return info